Description of problem: The clusteroperator marketplace will remain the status of `failing` is "True" and shows the message " Message: Operator exited " even the marketplace works well after the cluster upgrade from cv: "4.1.0-0.nightly-2019-04-25-121505" to cv: "4.1.0-0.nightly-2019-04-28-064010". describe the operator after the upgrade finished. Version-Release number of selected component (if applicable): from cv: "4.1.0-0.nightly-2019-04-25-121505" to cv: "4.1.0-0.nightly-2019-04-28-064010" How reproducible: always Steps to Reproduce: 1.install the cluster by nightly build "4.1.0-0.nightly-2019-04-25-121505" 2.check the status of marketplace #oc describe clusteroperator marketplace 3.upgrade the cluster to cv:4.1.0-0.nightly-2019-04-28-064010 4.check the status of marketplace after the upgrade finished #oc describe clusteroperator marketplace Actual results: 1.the status of marketplace before the upgrade after step 2 # oc describe clusteroperators marketplace Name: marketplace Namespace: Labels: <none> Annotations: <none> API Version: config.openshift.io/v1 Kind: ClusterOperator Metadata: Creation Timestamp: 2019-04-28T04:24:57Z Generation: 1 Resource Version: 131008 Self Link: /apis/config.openshift.io/v1/clusteroperators/marketplace UID: 9482b6ae-696d-11e9-9dae-0ad96e525668 Spec: Status: Conditions: Last Transition Time: 2019-04-28T08:34:11Z Status: False Type: Progressing Last Transition Time: 2019-04-28T08:34:11Z Message: 4.1.0-0.nightly-2019-04-25-121505 is available Status: True Type: Available Last Transition Time: 2019-04-28T08:33:51Z Status: False Type: Failing Extension: <nil> Related Objects: <nil> Versions: Name: operator Version: 4.1.0-0.nightly-2019-04-25-121505 2.the status of marketplace after the upgrade after step 4 # oc describe clusteroperators marketplace Name: marketplace Namespace: Labels: <none> Annotations: <none> API Version: config.openshift.io/v1 Kind: ClusterOperator Metadata: Creation Timestamp: 2019-04-28T04:24:57Z Generation: 1 Resource Version: 1316944 Self Link: /apis/config.openshift.io/v1/clusteroperators/marketplace UID: 9482b6ae-696d-11e9-9dae-0ad96e525668 Spec: Status: Conditions: Last Transition Time: 2019-04-29T08:50:34Z Status: False Type: Progressing Last Transition Time: 2019-04-29T08:50:34Z Message: 4.1.0-0.nightly-2019-04-28-064010 is available Status: True Type: Available Last Transition Time: 2019-04-29T08:49:59Z Message: Operator exited Status: True Type: Failing Last Transition Time: 2019-04-29T08:50:14Z Status: False Type: Degraded Extension: <nil> Versions: Name: operator Version: 4.1.0-0.nightly-2019-04-28-064010 Events: <none> Expected results: 1.The "Failing" 's status should be "False" after the upgrade. Additional info:
As I mentioned earlier, there's not really a use case for this. The new version of the API isn't aware of this field and won't do anything with it. So since we can't update it, and it won't be used, there's not much for the marketplace-operator to do in this case. My suggestion is that this bug should be closed as there's no op here.
There is no upgrade supported for this (Failed -> Degraded). WONTFIX this at this time.